arXiv subjects

Mohammad Golgol

Publications and source records attributed to Mohammad Golgol.

4 recordsLinked to original sources

arXiv2026

Impact of Work Schedule Flexibility on EV Hosting Capacity: Insights from Analyzing Field Data

Uncoordinated electric vehicle (EV) charging is altering residential load patterns and pushing distribution transformers to operate beyond their limits. These outcomes can be offset by exploiting the flexibility in work schedules (hybrid, remote vs. in-person) of EV owners, particularly when combined with rooftop photovoltaic (PV) generation. However, this phenomenon has not been explored in-depth yet. This paper addresses this research gap by introducing weekly work schedule-aware robust and chance-constrained optimization formulations for EV charging coordination to determine a transformer's EV hosting capacity. The results obtained using data from a residential feeder in Arizona indicate that an intelligent combination of work schedule flexibility with PV generation can help power utilities effectively manage changing grid demands.

eess.SY↗

arXiv2025

Exploring the Influence of Residential Electric Vehicle Charging on Distribution System Hosting Capacity -- A Case-Study in Arizona

The installation of high-capacity fast chargers for electric vehicles (EVs) is posing a significant risk to the distribution grid as the increased demand from widespread residential EV charging could exceed the technical limits of the distribution system. Addressing this issue is critical, given that current infrastructure upgrades to enhance EV hosting capacity are both costly and time-consuming. Moreover, the inherent uncertainties associated with EV charging parameters make it challenging for power utilities to accurately assess the impact of EVs added to specific locations. To address these knowledge gaps, this study (a) introduces an algorithm to coordinate residential EV charging, and (b) proposes a comprehensive framework that evaluates all transformers within a feeder. The proposed method is applied to a real-world feeder, which includes 120 transformers of varying capacities. The results demonstrate that this approach effectively manages a substantial number of EVs without overloading any of the transformers, while also pinpointing locations that must be prioritized for future upgrades. This framework can serve as a valuable reference for utilities when conducting distribution system evaluations for supporting the growing EV penetration.

eess.SY↗

arXiv2025

Topology-Aware Graph Neural Network-based State Estimation for PMU-Unobservable Power Systems

Traditional optimization-based techniques for time-synchronized state estimation (SE) often suffer from high online computational burden, limited phasor measurement unit (PMU) coverage, and presence of non-Gaussian measurement noise. Although conventional learning-based models have been developed to overcome these challenges, they are negatively impacted by topology changes and real-time data loss. This paper proposes a novel deep geometric learning approach based on graph neural networks (GNNs) to estimate the states of PMU-unobservable power systems. The proposed approach combines graph convolution and multi-head graph attention layers inside a customized end-to-end learning framework to handle topology changes and real-time data loss. An upper bound on SE error as a function of topology change is also derived. Experimental results for different test systems demonstrate superiority of the proposed customized GNN-SE (CGNN-SE) over traditional optimization-based techniques as well as conventional learning-based models in presence of topology changes, PMU failures, bad data, non-Gaussian measurement noise, and large system implementation.

eess.SY↗

arXiv2024

High-Speed Voltage Control in Active Distribution Systems with Smart Inverter Coordination and Deep Reinforcement Learning

The increasing penetration of renewable energy resources in distribution systems necessitates high-speed monitoring and control of voltage for ensuring reliable system operation. However, existing voltage control algorithms often make simplifying assumptions in their formulation, such as real-time availability of smart meter measurements (for monitoring), or real-time knowledge of every power injection information(for control).This paper leverages the recent advances made in highspeed state estimation for real-time unobservable distribution systems to formulate a deep reinforcement learning-based control algorithm that utilizes the state estimates alone to control the voltage of the entire system. The results obtained for a modified (renewable-rich) IEEE34-nodedistributionfeeder indicate that the proposed approach excels in monitoring and controlling voltage of active distribution systems.

eess.SY↗